Ever since reading over this thread, I’ve been wondering about the safety of my protected Li-Ion cells. I am especially interested in the low voltage protection that the PCB is supposed to offer. I believe that low voltage is the most probable danger with modern Li-Ion flashlights that often have a “soft” electronic switch that causes parasitic drain of the battery. Forgetting about the batteries being installed in such a flashlight could easily drain them down to 0V if there is no protection. Likewise, running a Li-Ion light on moonlight mode for a long time could also result in its voltage dipping into unsafe levels.
So, I finally got a DMM and I tested some Panasonic 3400 mAh protected button top cells from BangGood, and an XTAR 18700 cell that I got a few years ago. I tested the low voltage protection by first of all sticking a cell in a Uniquefire UF-X8 (XM-L) light in 1x18650 configuration on high and letting it run until the host flashlight’s own low voltage protection kicked in and shut it off. I chose this light because it has good heat sinking and won’t get ruined by running it on high for too long. But once the host’s low voltage protection kicked in at around 2.8V, I transferred the cell to a Solarforce L2P with an XM-L dropin that has no low voltage protection.
The good news: The two Panasonic cells from BangGood that I tested got down to slightly under 2.5V in the Solarforce host, which still let the the emitter give off a dim glow somewhere under 1 lumen or so. Slightly thereafter, the light shut off completely, and the cells tested at 0V, which I assume must be the protection circuit.
The bad news: The XTAR 18700 I tested was giving off a dim glow at slightly under 2.5V. But after leaving it on for a few more hours it got extremely dim, and eventually I couldn’t see any light with my eye. I thought the protection had finally tripped. But unfortunately my DMM actually measured it at 0.25V! Far too low. So it looks like at least on this cell the PCB is no good. I’ll have to also test the other XTAR 18700 I have to see how it behaves. EDIT: I tested the other XTAR 18700 I have, and it behaved exactly the same. Was very dim for a while at slightly under 2.5V, and then appeared to shut off, but it was still measuring at 0.18V. It acts like the circuit is not completely tripping, but rather artificially reducing the voltage to near 0.
Something i have noticed too, the protection circuits on some batteries seem to work fine or better with regulated lights than with a direct drive light.
I’ve found that with direct drive the light will often just get dimmer & dimmer far surpassing the supposed cut off protection value of the protection circuit.
Maybe the circuits have a design flaw that doesn’t trip under very low current draw circumstances.
Or maybe we’ve just got crappy protection circuits on some of our batteries.
This should get interesting when the knowledgeable folk chime in.
Instead of laptop pulls I get high drain cells from 18V Makita packs. I checked one that has been in the Ray gun since August and it still read 4.15V and that after 2 years use in power tools. Knock off packs don’t bother to wire in the cell monitoring circuits just the B+ and B- so it doesn’t surprise me to hear of disfunctional protection circuits. Thanks for the notification.
I have two new Trustfire 14500s that I got for some fairy lights, drawing around 50mA doesn’t activate the protection. Both cells discharged to below 2.5V. >.< The protection does work at higher current draw though.
Yeah, most PCMs have a minimum current needed before they will cut-off. With direct drive lights the current often drops so low that by the time the voltage gets down to the cut-off point it’s below that threshold and the PCM misses it completely.
I can’t understand how a direct drive light could discharge a cell all the way down to 0.25V though, as once the cells voltage drops below the Vf of the LED at minimum current it simply shouldn’t light up anymore. That would be about 2.6v for an XM-L iirc.
They are designed to take 3xAA alkalines. They’re pretty power hungry so I thought I’d try li-ion instead of eneloops. On 2x eneloops are just a bit too dim (they get more yellowish at lower voltage too) but last much longer. I was going to replace the “driver” with a step-down module but didn’t want to run it on 3x eneloops just to do so.
EDIT: I tested the other XTAR 18700 I have, and it behaved exactly the same. Was very dim for a while at slightly under 2.5V, and then appeared to shut off, but it was still measuring at 0.18V. It acts like the circuit is not completely tripping, but rather artificially reducing the voltage to near 0.
I applaud you for this testing sb. It makes one wonder about all the other batteries that have received a good rap but the circuit boards not tested as such. Thanks.
Are there ones that will work at <3V? I’d rather it didn’t make noise though. I have a little LED display which I might use just as an indicator, and use a larger cell so I don’t have to charge so often.
I tried two protection circuits from old Samsung cellphone batteries (~900mAh LiPo) and neither worked as required in this low current application. I was hoping they would at least prevent drain when a phone was turned off but it seems they don’t. I didn’t want to go beyond 2.7V though.
1s lipo low voltage alarms aren’t as common but they are out there. Red led at 3.3v for this one. Also there are some circuit schematics around for diy ones. Perhaps a driver with low voltage protection?
I hope it helps somebody to make an informed buying choice. Of course, as was mentioned above, a test set of qty. 2 isn’t exactly scientific evidence that all XTAR protection circuits behave this way, but at any rate it would be good to use extra caution.
